Extended Follow-up of Patients With Macular Edema Due to Retinal Vein Occlusion: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Extended Follow-up of Patients With Macular Edema Due to Retinal Vein Occlusion (NCT01198327), a na phase 3 study. The primary outcome is Incidence of Serious Adverse Events..

2. Clinical Question

The question is primarily about long-term safety in patients who had already received treatment in earlier studies. It is not a question about the effect of ranibizumab compared with an untreated group or an alternative treatment group. Because there is no concurrent comparator, events recorded during follow-up cannot by themselves establish how often similar events would have occurred in the absence of the study treatment or under another care strategy.

The record also names mean changes in visual acuity and retinal thickness as secondary outcomes. It does not include numerical results for those outcomes in the registry record for this analysis. The primary outcome is specified as the number of serious adverse events, rather than a comparative measure of treatment effect. The interpretation therefore centers on what the registry records about safety events and the limits imposed by a single-group follow-up design.

3. Trial Design

This is a phase 3, completed, single-group study with an enrollment of 66. The registry gives a start date of 2010-08 and a primary completion date of 2012-12. The study is titled as an extended follow-up of patients with macular edema due to retinal vein occlusion, including branch or central retinal vein occlusion, previously treated with intravitreal ranibizumab.

The brief description states that the study evaluates long-term safety among patients originally enrolled in BRAVO and CRUISE and subsequently followed in HORIZON. The treatment group is described as receiving ranibizumab as needed. The regimen specifies a 0.5 mg intravitreal dose, with treatments no less than 21 days apart, and optional peripheral laser to areas of non-perfusion. The registry lists masking as none and allocation as NA.

There is no randomized comparator in this record. The group therefore supplies follow-up observations, not a like-for-like control comparison. Prior participation in other trials is part of the study context, but details of how patients entered this extended follow-up, their earlier treatment assignments, or their treatment histories are not reported in the registry record.

The primary outcome is the incidence of serious adverse events, with a timeframe of 24 mos. The description says that ocular and non-ocular serious adverse events were recorded to gather long-term safety data. Mean changes in visual acuity and retinal thickness are listed as secondary outcomes. The registry record does not report a statistical analysis plan, an explanation of how the enrollment target was chosen, or details of the analysis populations. The sponsor is listed as Peter A Campochiaro, MD.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Incidence of Serious Adverse Events., measured over 24 mos.

The primary outcome is the incidence of serious adverse events, described as the number of serious adverse events during 24 mos. The record says the events include both ocular and non-ocular events and are collected to gather long-term safety data. This type of outcome is used to characterize serious harms observed during follow-up, rather than to measure a change in vision or retinal thickness.

The distinction between a count of events and a count of affected participants matters. One participant may experience more than one event, so an event count and a count of participants with at least one event are not necessarily interchangeable. The posted adverse-event figures include an affected field for the serious-event category, but the registry record does not give a separate total count of serious events in the registry record for this analysis. The affected figure should therefore not be presented as the total number of events.

The record names mean changes in visual acuity and retinal thickness as secondary outcomes, but does not give numerical results for them. Nor does it report details such as event adjudication procedures, severity grading beyond the serious-event classification, or the precise method used to summarize repeated events. Those details would affect how fully the outcome can be interpreted.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design allow?

In general, a single-group study records outcomes among participants who receive or have received the study intervention, without a concurrent control group. It can describe what was observed in the group, such as adverse events during follow-up. It cannot directly distinguish the intervention’s contribution from other explanations for those outcomes. A randomized comparison is often used to balance competing explanations between groups; this study is not described as having such a comparison. The record lists allocation as NA and the design as single group.

Here, that design is consistent with an extended safety follow-up among patients who had previously participated in BRAVO, CRUISE, and HORIZON. The study may provide information about events recorded during follow-up, but the registry record does not establish how observed event frequencies compare with what would have occurred without ranibizumab or with another treatment. A within-group description and a comparative treatment-effect estimate answer different questions.

How should an affected count be distinguished from an event count?

In general, an affected count represents the number of participants recorded as experiencing an outcome, while an event count represents the number of occurrences. If a participant has repeated events, the event total can exceed the number of affected participants. The registry lists an at-risk figure and an affected figure for serious and other adverse events. It also defines the primary outcome as a number of serious adverse events. The separate event total and the method used to handle repeat events are not reported in the registry record, so the affected figures should not be relabeled as total event counts.

This distinction is important for safety reporting. A participant-level summary can indicate how many people experienced at least one event, while an event-level summary can reflect repeated occurrences. Those summaries have different interpretations and may require different statistical methods. Without a documented definition of the displayed fields and any repeat-event rules, interpretation should remain close to the wording used in the registry.

What does a confidence interval usually communicate?

In general, a confidence interval describes statistical uncertainty around an estimated quantity under the assumptions of the method used. A narrower interval indicates greater precision under those assumptions; a wider interval indicates less precision. A confidence interval is not a guarantee that the true value lies within its bounds, and it does not remove limitations such as lack of a comparator or incomplete outcome reporting. No confidence interval for the primary safety outcome is reported in the registry entries available for this analysis, so precision cannot be assessed from an interval.

For event outcomes, the appropriate interval depends on the target quantity and the analysis method. A participant-level proportion, an event rate over person-time, and a count of recurrent events are not interchangeable estimands. The registry record does not report person-time, an event-rate calculation, or a model for repeated events. It is therefore not possible to identify which inferential approach, if any, was used to quantify uncertainty.

Why can over-dispersion matter?

In general, count models may assume a particular relationship between the mean and variability of counts. Over-dispersion occurs when the observed variability is greater than a model assumes. If it is ignored, standard errors may be understated and confidence intervals may be too narrow. Over-dispersion can arise when event counts vary more across participants than the chosen model allows. This is a general statistical issue for event-count analyses; the registry record does not report a count model, a dispersion assessment, or a model-based estimate for this outcome.

Repeated events also create dependence: events from the same participant are not necessarily independent observations. General methods for recurrent events account for this structure in different ways. The registry record does not describe whether repeated events were present or how any such dependence was handled. No specific model should be attributed to the study on the basis of the registry entries available.

How should planning illustrations be read?

In general, sample-size calculations depend on a target effect, variability assumptions, a significance level, allocation, and a statistical model. The planning assumptions associated with the calculations described for this record specify a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect set as a stated fraction of the observed effect. They are planning illustrations computed from the observed effect, not a statement of this study’s original design.

That distinction is especially important because the clinical study is single group, whereas the planning assumptions describe equal allocation between two arms. Such calculations do not turn the study into a two-arm comparison, and they do not indicate the study’s observed power. They are hypothetical planning quantities under stated assumptions. The registry record does not report the original sample-size justification or say that those assumptions guided trial conduct.

More generally, power is a design property tied to a specified hypothesis and assumed effect under a planned analysis. It should not be calculated after observing results and then used to judge whether the observed result is credible. A planning illustration based on an observed effect can instead show how a future design might be sized under explicit assumptions. It remains sensitive to the choice of effect fraction and model assumptions, and it does not replace reporting the actual estimate and its uncertainty.

The record states that time-to-event planning calculations use total events, while other calculations use participants per arm. That general distinction reflects different design quantities: time-to-event planning often depends on the number of outcome events, whereas other designs may be expressed in participants per group. No particular planning result should be taken as a feature of the completed single-group study.

10. Multiplicity, Missing Data and Analysis Populations

The record lists the primary outcome as serious adverse events and names mean changes in visual acuity and retinal thickness as secondary outcomes. It reports three outcome measures overall, while the named secondary outcomes account for two listed outcome types. The registry record does not report a multiplicity strategy, such as a hierarchy of hypotheses, an adjustment to significance thresholds, or a plan for controlling the chance of false-positive findings across outcomes. Without such information, statistical significance across multiple outcomes cannot be interpreted as though a prespecified multiplicity method had been used.

The registry entries available for this analysis do not include numerical results for the secondary outcomes or inferential statistics for the primary outcome. The record also reports zero analyses. A statistical analysis plan, specific analysis populations, and methods for handling missing outcome data are not reported in the registry record. No assumptions about imputation, exclusions, or follow-up completeness can therefore be made.

Missing data can affect both the number of participants contributing information and the meaning of an estimate. In general, the consequences depend on why data are missing and how the analysis handles them. For safety outcomes, incomplete follow-up may also affect the opportunity to observe events. The record does not provide enough information to evaluate these issues, including the extent or pattern of missing data.

Because this is a single-group follow-up, the absence of an analysis plan also limits evaluation of how outcomes were summarized and whether any covariate adjustment or subgroup analysis was intended. Such methods are not reported in the registry record and should not be presumed.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hospitalized for pneumonia (3.0%), hospitalized for elective surgery (3.0%), death resulting from pre-existing idiopathic pulmonary fibrosis (1.5%), death of unknown cause (1.5%), ecg for elective knee surgery showed inverted t waves, patient asymptomatic, admitted for workup (1.5%).

The registry lists 66 at risk and 17 affected for serious adverse events, and 66 at risk and 53 affected for other adverse events. These figures describe the affected fields, not necessarily the number of separate event occurrences. The primary outcome is worded as a number of serious adverse events, but a distinct total event count is not reported in the registry entries available for this analysis.

Several serious-event terms are listed with percentages: hospitalization for pneumonia and hospitalization for elective surgery are each listed as 3.0%; death resulting from pre-existing idiopathic pulmonary fibrosis and death of unknown cause are each listed as 1.5%; an ECG finding during elective knee surgery and mild chest discomfort with coronary angiography findings are each listed as 1.5%. These entries are descriptions, not evidence of a causal link to study treatment. The record does not specify whether terms overlap or whether the list is exhaustive.

Deaths appear among the listed serious-event descriptions, but the death field is null. The event descriptions and percentages should therefore not be used to infer an overall death count. The single-group design also means the percentages do not establish comparative risk.

12. What the Result Does, and Does Not, Mean

The record documents a completed, single-group extended follow-up focused on serious adverse events in patients with macular edema due to retinal vein occlusion who had previously participated in other studies. It describes the treatment regimen and reports affected-field figures for serious and other adverse events, along with selected serious-event terms and percentages. This is useful as a descriptive account of the safety information recorded for the follow-up group.

It does not establish that ranibizumab caused a listed event, prevented an event, or changed risk relative to another treatment. There is no concurrent comparator, and the listed event terms do not supply a comparison with an untreated or differently treated group. The record also does not report the context needed to estimate comparative risk, such as an analysis plan, a separate serious-event count, or a time-at-risk analysis.

The registry lists visual acuity and retinal thickness as secondary outcomes, but numerical results for these outcomes are not reported in the registry record for this analysis. Consequently, the safety information cannot be used to infer improvement or worsening in either outcome. Nor does the record report a confidence interval for the primary endpoint, so the statistical precision of the safety estimates cannot be judged from the information available.

The findings apply most directly to the group described: patients previously enrolled in BRAVO or CRUISE and subsequently followed in HORIZON. The registry record does not report enough information to assess how closely this group represents patients with retinal vein occlusion in other settings. Overall, the appropriate interpretation is descriptive and specific to the recorded follow-up, not a definitive comparative assessment of treatment benefit or harm.

13. Why This Trial Matters Statistically

Extended follow-up addresses a different statistical question from a short-term treatment comparison. It can document safety events recorded over a longer observation period among people already exposed to treatment. That information can be relevant even when the study is not designed to estimate a comparative effect. The key is to keep the descriptive question separate from a causal comparison.

This study also illustrates why registry outcome wording matters. The primary outcome is described as a number of serious adverse events, while the adverse-event fields report numbers at risk and affected. These may represent different quantities: event occurrences and participants affected are not interchangeable. Clear definitions are important for readers, analysts, and future evidence syntheses.

The single-group design and the absence of reported uncertainty estimates make restraint especially important. An observed count describes what was recorded, but without a control group it does not indicate how the event experience differs from another strategy. Without a confidence interval or stated analysis method, readers also lack a statistical account of precision. The findings should not be made more conclusive than the record supports.

Finally, the planning assumptions described for sample-size illustrations are not the same as the study’s design. They include equal allocation between two arms, while the clinical record is a single-group follow-up. Recognizing this distinction prevents hypothetical planning quantities from being mistaken for design features or observed power. Taken together, these issues show how design, endpoint definition, analysis reporting, and uncertainty shape what a safety record can answer.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record does not report a concurrent comparator, so the safety findings are descriptive rather than a direct estimate of relative risk. It also does not report a separate total count of serious adverse events, even though the primary outcome is worded as a number of events. The affected-field figure cannot be assumed to be an event count.

The record does not give the method for identifying, classifying, or adjudicating serious events, or explain how repeat events in the same participant were handled. It does not report person-time or a time-to-event analysis. These omissions limit interpretation of how event frequency was summarized over the 24 mos timeframe.

The group consists of patients previously enrolled in BRAVO or CRUISE and subsequently followed in HORIZON. Details of selection into this extension, participant characteristics, earlier treatment assignments, and completeness of follow-up are not reported in the registry record. Generalizability beyond this group cannot be assessed from these registry entries alone.

The registry lists a null death field while also including death descriptions among the serious-event terms. It does not report whether the terms overlap or constitute an exhaustive list. The listed percentages therefore should not be treated as a complete accounting of deaths or serious events. Numerical results for the secondary outcomes, a statistical analysis plan, missing-data methods, and outcome-specific uncertainty estimates are also not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Extended Follow-up of Patients With Macular Edema Due to Retinal Vein Occlusion — NCT01198327.
• Registry record last updated 2014-01-13; status last verified 2013-11.
• Results first posted 2014-01-13.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

The record concerns a completed phase 3, single-group extended follow-up study of patients with macular edema due to retinal vein occlusion who had previously participated in BRAVO or CRUISE and were followed in HORIZON. Enrollment is listed as 66. The treatment description specifies as-needed intravitreal ranibizumab at a 0.5 mg dose, no less than 21 days apart, with optional peripheral laser to areas of non-perfusion.

The primary outcome is the number of serious adverse events, ocular and non-ocular, over 24 mos. The registry lists affected-field figures for serious and other adverse events and selected serious-event terms. Mean changes in visual acuity and retinal thickness are named as secondary outcomes, but numerical results for those outcomes are not reported in the registry record for this analysis.

There is no comparator group. The registry record does not report the analysis plan, missing-data approach, or a separate total event count. The evidence is therefore best read as descriptive safety follow-up rather than a comparative estimate of treatment effect.